Recombination between the co-packaged genomes of HIV-1 is known to produce more potent virus that frustrates immune response and attempts at therapy. Recombination occurs during the reverse transcription steps of HIV-1 replication, with one mechanism involving the transfer of the growing DNA primer from one viral RNA template to the other. Our work is focused on determining the mechanisms that result in recombination during minus strand synthesis. Earlier we advanced evidence that pausing of the reverse transcriptase during synthesis concentrates RNase H directed cutting of the original template. This clears an area for interaction or invasion by the second template that promotes transfer. The process is also enhanced by template-template interactions. More recently we discovered that transfers occur in two distinct steps. After the initial interaction of the second template with the DNA, the first step, the DNA-RNA hybrid region propagates to a downstream site where the DNA 3' terminus transfers, the second step. We propose to determine distinct properties of the reverse transcriptase (RT) and the template structure that promote the second step of transfer. Mutant RTs and altered template structures will also be used to determine the reasons for the previously observed time delay in transfer, and the dynamics with which the propagating hybrid catches the terminus. We have evidence that the transfer of minus strong stop DNA during HIV-1 replication also occurs by a two-step mechanism. It is inefficient when reconstituted with short templates in vitro. However, longer templates show similar efficiency to that in vivo. We will explore how specific long distance interactions within the templates both up- and downstream of the transfer site promote transfers. We have developed an HIV-1 cell culture system designed to determine whether transfer mechanisms observed in vitro occur in the same manner in vivo. Use of RT mutants and viral genomes with specific alterations in template structure will be used for this task. Lastly, recent evidence suggests that cleavage specificity of the RT RNase H is a critical factor in determining the efficiency of transfer reactions. Specifically important is the ability of the RT to make adjacent cuts, by a mechanism of primary-secondary cuts that we previously described. Use of specific RNase H mutants of RT also holds the promise of determining the role of other features of RNase H specificity.